Bad idle when its cold.. fast idle when its hot.

I've made a lot of progress on diagnosing and I want to combine them all into one thread so I can get this figured.

Here's what I got:
1989 Cherokee Pioneer 4.0L Inline Six RENIX controlled, Manual Peugeut BA/10 transmission.

Here's what's happening:
Anytime I start it, it revs up in between 1500-2000 RPM's then settles down to the idle it wants.

When it's cold, it'll idle at 650 RPM's. If I give it any throttle in neutral, it will hesitate (NOT backfire) and then rpm's go up. If my RPM's are 1000'ish+ and I let off the throttle alltogether, my RPM's will dip below the 500 mark, sit there at a very rough almost-dying idle for a few seconds, then either A) stall, or B) pick up to 650ish RPM's.

When it's warming up, the idle will get progressively faster. I think it goes up 50 rpm's per minute of idling.

When it's hot, it will idle at 1000 RPM's, and have absolutely no hesitation issues whatsoever, and RPM's only dip 50 below idle for a second if I rev it high and let off the gas.

After it's hot and I'm driving it normally (light foot, shifting at 2000 rpm's etc), my idle will change to 1300 RPM's when I put the clutch pedal in or be in neutral. IF I really quickly tap on the pedal, RPM's might settle down to 1000. They will pick right back up again after driving. IF I shut off the vehicle when it's idling at 1300, and then restart it right afterwards, it idles at 1000 until I start driving again.

Here's what I've done & diagnosed:
Bought new fuel filter, air filter, plugs, wires, cap, rotor, O2 sensor, TB Gasket. - No effect.

Changed TPS with another I had laying around (Dunno where it was from) - No effect.

All top intake manifold bolts were tight, and 3 of the bottom ones I could find were tight.

Renix manual says that the 4-plug connector underneath distributor should get an input of 5.0 volts. Mine gets an input of ~7 volts.

Redid my entire ground setup - No noticable effect.

TPS output is at 0.79 volts
TPS input is 4.93 volts
When idling at 1000, MAP sensor terminals A&B voltage is 1.2-1.3 volts
MAP sensor input voltage is 5.05 volts
IAT & CTS sensors seem on par with charts.
Crankshaft Position Sensor resistance is good.

Knock sensor is unplugged.

EGR valve rod looks fully extended (Couldn't move it anymore with needle nose pliars)

Shop I took it to couldn't find any vacuum leaks with carb cleaner.
I sprayed everything and couldn't find any leaks. I did notice a very very subtle RPM increase by spraying TB cleaner on the EGR rod - it might have been coincidince because continuous spraying didn't really affect it.

I don't really know where to go from here. Is my ECU screwed up? I doubt it. Could my IAC be screwed up? TPS? Thanks in advance, everyone.

It sounds like a vacuum leak to me - there are about six screws on the bottom of the manifold, but the back ones are difficult to reach. Getting a 9/16" socket with an inbuilt universal joint will help you reach them, believe me!

Also, check to make sure your EGR tube is tight (especially the intake end,) the EGR mounting screws are tight, and the throttle body and any other metal-on-metal (with or without gasket) seal that holds vacuum.

What gives me the "vacuum leak" idea is that it settles down and goes away when your engine is warmed up - and that would be why I'm telling you to check any metal-on-metal seal, since metal expands as it is heated and that will cause the seal to be restored (especially when aluminum is involved.)

Torque specs for just about everyting RENIX are on my site in the Tech Archive, so check there for anything you don't already have. They're compiled from the FSMs, so I'm fairly sure of them (especially since I've used most of them myself.) There are a couple of things I want to make notes on - and I'll be doing that as I'm updating other things.

Bear in mind that most of the cues with the carburettor cleaner trick are visual - and that makes it likely to miss things, if you're not careful. It's a good tip for finding vacuum leaks, but it's not a panacea for finding them...

How easy it is to forget! I had EGR bolt problems myself 2 years ago, had to re thread one of mine when I replaced my EGR valve and discovered a partly stripped female thread.

Alrighty I'll tighten em up.

Could a shot brake booster cause some of these problems?

I've been noticing that my brakes have always sucked on this jeep. Always. The front rotors and pads are pretty much brand new, and the back shoes/cylinders I did last weekend. They still suck.

Also, after sitting overnight, a brake booster should have vacuum still right? Well without starting the engine, I put on my brakes and low and behold, its as if I dont have a booster at all. It's completley stiff and I almost back into my trailer. I'll try plugging me brake booster line and see what happens.

Check the vacuum check valve and grommet - that could also cause a vacuum leak (you are correct.)

If you have decent brakes when you've been running or idling for a while, start with the check valve and grommed on the front of the thing. If not, you may have a pinhole in the diaphragm inside...

I have absolutely crap crap crap brakes when its been sitting and its not running, and after its been running for awhile, my brakes are still pretty bad. I dont have ABS, and when I absolutely slam on my brakes, I dont do any locking up. Its also really stiff at the bottom.

Well I disconnected the brake booster hose from the booster and put a wooden dowel in there, perfect fit, annd all my problems are still very evident..

mattbred said:

I have absolutely crap crap crap brakes when its been sitting and its not running, and after its been running for awhile, my brakes are still pretty bad. I dont have ABS, and when I absolutely slam on my brakes, I dont do any locking up. Its also really stiff at the bottom.

Click to expand...

This is related only to the brakes, there is a proportioning valve right after the master cylinder that can get clogged up, and cause poor brake performance.

And yes, a vac leak at the brake booster diaphragm or hose or check valve can cause wierd idle varations too.

Well it could be my calipers, or my booster. going to check it out this weekend. A junkyard booster wouldn't hurt.

I noticed that when it hesitates upon acceleration, my vacuum pressure goes to zero, absolute zero. Also my air intake really makes a slurping sound. TB does it when air intake is disconnected.

Also, the guy that did the fuel pressure test noted that it was 39 when it was not running, and 30 when it was running (didnt check fuel regulator).

Just puttin that out there.

Update: All manifold bolts are tight. Took a wrench to them all (son of a !@#$ that was hard). I took the entire EGR system out, and will cover it up with a plate. Who needs it anyway!

If it's not the EGR, or intake manifold, then I'm going to buy a new IAC. If it's not the IAC.. Then I'll get my ECU tested I guess. Eugh!

Check vacuum at your brake booster port, then find a way to "T" in and check vacuum again with the booster connected. It's starting to sound like yours is leaking, and I'd be a bit leery about using a JY booster without going through it first - that's an awfully critical part...

Well I checked vacuum at a different port, both with and without the booster connected, and it seemed the same. If I could T it I would.

Also what do you mean about using a JY booster without going through it? I dont get what you mean.

Edit: Ooh a JunkYard booster. I see. Yeah maybe a vacuum pump would be a good idea to take along with me.

I was experiencing the same thing with my '89 ... exactly. Idle that was all over the map and brakes that are weak and downright squirrely especially under a hard brake.

The idle problem went away when it occurred to me that I was actually putting my foot into the gas pedal a little when starting it. It seemed to mess up the idle sensor and at times it would idle even as high as 1600 rpm (made shifting gears interesting). No stomping of the pedal or pleading would correct it until I turned it off for 10 minutes and restarted it. Once I stopped doing that and just let the injectors do their thing when starting the problem has not resurfaced. It may be just fluke, but it seems on my Jeep at least, the crappy brakes and the idle problem seem on the surface to be unrelated. I do see the logic in the braking vacuum theory though.

I was planning on replacing the entire brake system, drums, rotors, pads, shoes, cylinders and booster so I know the whole thing has been redone and is at least safe. I somehow feel a little better knowing this is common on the older XJs and not just a crappy Jeep over here.

Good luck!

Update:

I took out the EGR and made meself a blockoff plate with gasket material and a piece of tin, and it went in really easily. Started up me engine and revved a bit.. Absolutley no difference. I revved it to 2000 RPM's, it dropped down to 400ish RPM's, starting shaking and rattling, revved up a tiny bit to more stable, then back down to the 0 mark on the tach for two seconds before I turned the key off.

Another interesting thing happend today. It started up and settled to an idle at 500 rpm's, and it was a rough idle. Shaking a bit, then revving up by 50-100, then down and shaking a bit. Sometimes itd go lower than 500 and start to almost-stall for a few seconds then go back up to 500ish.

I took off a vacuum port, and idle didnt change much, and then I took an unlit propane torch to the port and idle didnt change at all with the influx of propane. If anything it went down. Does that mean its rich? Also I wiggled some fuel injectors, and one of them leaked a bit of fuel onto my intake manifold when I wiggled it.

You need to pay close attention to where the injector(s-?) leaked or is leaking. Three possibilities, the intake maniflod/injector o'ring, the fuel rail o'ring, both are easy fixes, just replace all the o'rings, about a 60 minute job.

Third is the injector body itself near the electrical connection of the body or center of the body, in which case they are probably OEM injectors which are notorious for body leaks and they are a fire hazard. I went through about 18 JY OEM injectors and gave up on them, bought a new matched flow set of superior injectors from fiveomotorsports.com for less than $250 as I recall. Others here swear by 19lb Ford JY injectors (I think). Do a search on fuel injectors for details.

You can use a rubber vacuum hose of stethoscope to listen to each injector to see if it is firing, it sounds like a lifter tick, tick, tick sound with the listening device. If one is not firing all the time or not at all you can here the differance.

How old are the injectors?

Propane question, sounds like it is running rich, but

Idle not changing much,......means IAC may be working?

mattbred said:

Another interesting thing happend today. It started up and settled to an idle at 500 rpm's, and it was a rough idle. Shaking a bit, then revving up by 50-100, then down and shaking a bit. Sometimes itd go lower than 500 and start to almost-stall for a few seconds then go back up to 500ish.

I took off a vacuum port, and idle didnt change much, and then I took an unlit propane torch to the port and idle didnt change at all with the influx of propane. If anything it went down. Does that mean its rich? Also I wiggled some fuel injectors, and one of them leaked a bit of fuel onto my intake manifold when I wiggled it.

Click to expand...

AHA! I think I fixed part A of my high idle. I sprayed a shit load of TB cleaner through the idle-port, and also with the EGR replaced with tin and a home made gasket, my hot non-revving HOT idle is 750 instead of 1000! VICTORY!

Although I do have a COLD low idle now (vacuum leak musta made it higher at cold startup) of 500 RPM's, with bad hesitation and coming back from rev-stalling. Do NOT know what that is.

Part B of my high idle situation is still there, yet not as intense since my default idle is lower. When it's hot and I rev it, my idle does go up to 800-1000, so I'm thinking that's the TPS. Well, actually I will tell you about that.

My TPS resistance, NOT connected to anything, remains absolutely rock-stable. For example (not the real resistance, but for an example), with TPS not connected to anything but the ohms meter, it would give a resistance of say, 6000 ohms (NOT exact just example). WOT would be 4000 ohms. Now, rocking between it slowly and fastly would always come back to either 6000 or 4000. NO problem.

HOWEVER when I connect the TPS to the wiring harness and check resistance, itll be say 2800 at idle, and 2200 at WOT. When I rock the TPS slider back and forth, it will NOT be exactly 2800 and NOT be exactly 2200. It'll be different depending on how fast I move the TPS slider. I can't explain it very well, so I hope you understand.

Simpler: TPS not connected to harness, resistance is rock stable. TPS connected to harness, resistance varies.

Check each wiring harness (TPS) connector (all six connectors) to ground. With the power OFF, some should read infinite resistance others should read less than one ohm. Any that read somewhere in between 1 ohm and infinite are suspect!

Check each one while moving the wiring harness to see if it changes when the harness is moved. If not you may just be having trouble getting good repeatable clean connections between the test probes and assembled TPS to harness connectors. I had that problem and finally chose to make all my measurements on the seperated, disconnected components one at a time to ground on the ECU side. That is why I had to come up with my own data table since all the posted data and test methods were for the assembled system powered up.

How about taking and posting real ohms values?

Ecomike said:

Check each wiring harness (TPS) connector (all six connectors) to ground. With the power OFF, some should read infinite resistance others should read less than one ohm. Any that read somewhere in between 1 ohm and infinite are suspect!

Check each one while moving the wiring harness to see if it changes when the harness is moved. If not you may just be having trouble getting good repeatable clean connections between the test probes and assembled TPS to harness connectors. I had that problem and finally chose to make all my measurements on the seperated, disconnected components one at a time to ground on the ECU side. That is why I had to come up with my own data table since all the posted data and test methods were for the assembled system powered up.

How about taking and posting real ohms values?

Click to expand...

Sounds good! I'll take my notebook and take some values tomorrow. Damn broken ankle...

Good job you guy's on the advise! and tips, i learned a lot in my quest to trace down a high hot idle and a low cold idle problem. I also have an 89 and the brake booster does not seem to be working, the pedel is soft, goes down a long way before the brakes start to work. just replaced rear shoes, no big improvement.When you push down on the pedel shouldent you hear some vacume sound? How do you test a booster?